Prince William Sound Regional Citizens Advisory · PDF filePrince William Sound Regional...

Prince William Sound Regional Citizens Advisory Council

Valdez Marine Terminal Storage Storage Tank Mechanical Integrity Investigation

(Contract No. 505.2007.02)

for:

Dr. Thomas H. Kuckertz

PWSRCACP.O. Box 3089

Valdez, AK 99686

The Hendrix Group, Inc. Report No. H27053

by:

The Hendrix Group, Inc.15823 N. Barkers Landing

Houston, TX 77079(281) 556-8774

September 14, 2007

tkuckertz

Text Box

The opinions expressed in this PWSRCAC commissioned report are not necessarily those of PWSRCAC.

tkuckertz

Text Box

-a-

Table of Contents

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 1 of 22Tank 55 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . of 22Tank 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 2 of 22

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 3 of 22

Background Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 3 of 22

Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page4 of 22

Site Visit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 5 of 22Hardness testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 5 of 22Doorsheet Radiographic Film Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 6 of 22

Tank 55 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 6 of 22Tank Design and Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 6 of 22Base and weld metal properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 7 of 22Tank 55 Unsats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 7 of 22WPS/PQR’s (UNSAT #19) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 9 of 22No heat input monitoring for doorsheet weld (UNSAT #20) . . . . . . . . . . . Page 11 of 22Welding out of sequence on door sheet . . . . . . . . . . . . . . . . . . . . . . . . . . Page 14 of 22Fitness-for-Service Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 14 of 22Introduction to brittle fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 15 of 22Brittle Fracture Assessment to API 579, Section 3 . . . . . . . . . . . . . . . . . . Page 17 of 22Tank 55 Fitness-for-Service Opinion . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 18 of 22

Tank 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 18 of 22Tank Design and Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 18 of 22Base and Weld Metal Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 19 of 22Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 20 of 22Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 20 of 22Opinions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 21 of 22

Tank 55 and 16 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 21 of 22

Appendix A - References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 26

Appendix B - Tank 55 Photographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 32

Appendix C - Tank 55 Doorsheet Hardness Results . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 37

Appendix D - Tank 55 WPS Comparison Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 39

Appendix E - API 579 Brittle Fracture Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 41

Appendix F - Listing of Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 45

The Hendrix Group Inc. Client: PWSRCACHG report No. H27053 Valdez Marine Terminal Tank Issues

Page 1 of 22

EXECUTIVE SUMMARY

Prince William Sound Regional Citizens Advisory Council (PWSRCAC) requested The HendrixGroup, Inc.’s assistance to investigate the alleged tank integrity issues for Valdez MarineTerminal (VMT) Tank Number 55 and Tank Number 16. The tank mechanical integrity issueswere based on allegations of irregularities associated with repairs and inspections made to thetwo tanks by contractors and employees of the Alyeska Pipeline Service Company (APSC)during the 2002 time frame.

The efforts described in this report were a joint effort funded by Prince William Sound RegionalCitizens’ Advisory Council and Alyeska Pipeline Service Company. In addition to jointly fundingthe effort, Alyeska provided complete access, as and when requested, to its qualitydocumentation, personnel, and physical plant that greatly aided the exploration of the tankintegrity issues described herein.

Allegations associated with Tank 55 that were investigated included:

1. Door sheet welded with wrong welding procedure

2. No heat input monitoring for door sheet weld

3. Welding out of sequence on door sheet

Allegations associated with Tank 16 included:

A. Leaking of product under the tank floor.

This report summarizes the work completed by The Hendrix Group, Inc. in the course of theinvestigation into the issues related to Tanks 16 and 55. PWSRCAC requested that theinvestigation include specific tasks associated with the two tanks, including the following:

Task 1 – Review welding documentation, radiographic film and other quality documentationassociated with welds used to reinstall Tank 55 door sheet during 2002. Performon-site visual inspection of the welds. If appropriate, recommend and observenondestructive testing performed by APSC or contractor of the welds and adjacentheat-affected zones.

Task 2 – Determine if the welding procedures were prepared and qualified acceptably inaccordance with applicable standards (e.g., API 650, API 653, ASME Section IX),APSC procedures, federal and state regulations, and other applicable industrialcodes and practices. Assess the extent to which the procedures used wereequivalent and the extent to which the integrity of the welds made in accordancewith the procedure for the thicker material and of the adjacent heat affected zonesmay have been compromised. Assess the extent to which the design basis of thetank may have been compromised.


The Hendrix Group Inc.

15823 N. Barkers Landing Ph: 281.556.8774Houston, TX 77079 Fax: 281.870.0659

Page 2 of 22

Task 3 – Assess the extent to which existing in-process information is sufficient todemonstrate adherence to the essential variables and heat input requirements andlimitations of the welding procedures.

Task 4 -- Assess the impact of any variances to procedure essential variables (including heatinput) on weld properties, the significance of any variances on weld integrity andoverall concern for weld failure. Evaluate weld workmanship and acceptability withrespect to the radiographic examination acceptance criteria.

Task 5 – Prepare a written report summarizing the review, findings, and recommendations

The investigation results showed that:

Tank 55

• While certain irregularities occurred during installation of the doorsheet, those irregularitiesdid not influence the tank fitness for service, based on brittle fracture and weld integrityconcerns.

• While the original welding procedures approved for welding the Tank 55 doorsheet werenot in compliance with API Standard 650, Welded Steel Tanks for Oil Storage, and APSCproject specifications, revisions to the essential variables of the original procedures madeby APSC to correct the deficiencies (as documented by inspection UNSATS) were properand did not violate any code or project specification requirements.

• Welding with initially incorrect welding procedures versus welding to the revisions made tocorrect those procedures would not have materially altered the actual welding process anddid not compromise the doorsheet welds.

• Issues regarding the doorsheet welding procedures and the actual welds made did notcompromise the original tank design.

• Review of the available documents did not permit an unqualified verification as to whetherproject specification and welding procedure requirements pertaining to heat inputrequirements were met; however, the discussion will show that, based on the availableinformation and the line of thought conducted, any deviations from the requirements did notcompromise the quality or fitness -for-service of the doorsheet welds.

• An API Recommended Practice 579, Fitness-for-Service, Level 1 and Level 2 assessmentregarding the potential for brittle fracture showed that, based on the available information,Tank 55 passed the Level 1 and 2 assessments and is considered fit-for-service, based onbrittle fracture concerns.




Page 3 of 22

Tank 16

• It could not be verified whether Tank 16 ever leaked hydrocarbon during the 2002 timeperiod; however, an inspection of the exterior of the tank in June 2007 showed that it couldnot have been leaking significant quantities of product into the ground.

RECOMMENDATIONS

An observation that we made during the course of the VMT employee interview process atValdez and based on our document review were perceived deficiencies in ASPC’s abovegroundstorage tank mechanical integrity procedures and processes. An example of this arose whendiscussing whether APSC conducting a hydrotest of Tank 55 after the doorsheet replacement. APSC had apparently exempted the tank from a hydrotest, based on the exemption criteria inAPI 653; however, they did not have documentation of the exemption in the tank files nor didthey have an engineering review procedure for conducting a hydrotest exemption engineeringreview.

A review of documents submitted by APSC showed two procedures which appeared to apply toAPI 650 storage tanks: (1) X052-T-411 (rev. 0), “Aboveground Storage Tank Repair andAlteration and, (2) MP-166-3.20 (rev. 6), Integrity Management Engineering Monitoring ProgramProcedures. Both procedures referenced other APSC internal procedures and externaldocuments. MP-166-3.20 appears to be the governing procedure assuring compliance withState and Federal Regulations. Curiously, MP-166-3.20 references X052-T-411, but X052-T-411 does not reference MP-166-3.20. MP-166-3.20 is a high level document generally outliningan overall process but is short on lower level “how to” procedures, such as the above mentionedhydrostatic exemption issue.

As an example, other tank procedures that might prove useful to APSC would include:

(1) Requirements for the Evaluation of Tank Inspection Data,

(2) Requirements for the Evaluation of Tank Change-of-Service Requests,

(3) Procedure for the Development of an Inspection Company Approved Vendor List.

(4) Requirements for The Inspection of Fixed Roof, Atmospheric, Welded Storage Tanks.

(5) Requirements for The Inspection of Internal Floating Roof, Welded Storage Tanks.

(6) Requirements for The Inspection of External Floating Roof, Welded Storage Tanks.

(7) Tank hydrostatic test requirements and exemptions




Page 4 of 22

(8) Procedure for evaluating non-conforming tank inspection data.

(9) Procedure for granting tank next inspection interval extensions.

BACKGROUND INFORMATION

During the year 2002 various repairs and modifications were made to several abovegroundstorage tanks at the Valdez Marine Terminal (VMT). In May 2006 PWSRCAC (the Council)received a series of letters from a whistle blower that alleged faulty welds, incorrect use ofwelding procedures, and regulatory indifference pertaining to unspecified welding situations atthe Valdez Marine Terminal involving the 2002 tank repairs and modifications. The Council andAlyeska Pipeline Service Company (APSC) identified a set of welds used to reinstall a doorsheet in 2002 that had been cut to allow access to Tank 55 as, at least in part, the basis for theallegations. PWSRCAC wanted to verify the extent to which the Tank 55 door sheet welds mighthave compromised the mechanical integrity of Tank 55 by examining the procedures used tomake the welds, the standards and regulations pertaining to the welds, and the weldsthemselves.

This report addresses certain focused issues related to allegations of improper tank repairs forTank 55 and leaks associated with Tank 16. Specifically, it addresses the following “Unsat”numbers generated during the repairs.

UNSAT 19 - Shell thickness variable in weld procedure greater than actual. Requires newprocedure.

UNSAT 20 - Welding out of sequence on door sheet.

SCOPE

As requested by RCAC, the VMT investigation initially covered the following specific projecttasks related to Tank 55:

Task 1 – Review welding documentation, radiographic film and other quality documentation(Appendix A) associated with welds used to reinstall Tank 55 doorsheet during2002. Perform on-site visual inspection of the welds. If appropriate, recommendand observe nondestructive testing performed by APSC or contractor of the weldsand adjacent heat-affected zones.

Task 2 – Determine if the welding procedures were prepared and qualified acceptably inaccordance with applicable standards (e.g., API 650, API 653, ASME Section IX),APSC procedures, federal and state regulations, and other applicable industrialcodes and practices. Assess the extent to which the procedures used wereequivalent and the extent to which the integrity of the welds made in accordance




Page 5 of 22

with the procedure for the thicker material and of the adjacent heat affected zonesmight have been compromised. Assess the extent to which the design basis of thetank might have been compromised.

Task 3 – Assess the extent to which existing in-process information is sufficient todemonstrate adherence to the essential variables and heat input requirements andlimitations of the welding procedures.

Task 4 -- Assess the impact of any variances to procedure essential variables (including heatinput) on weld properties, the significance of any variances on weld integrity andoverall concern for weld failure. Evaluate weld workmanship and acceptability withrespect to the radiographic examination acceptance criteria.

Task 5 – Prepare a written report summarizing the review, findings, and recommendations.

During the course of the project, another task was added to the project scope involvinginvestigating certain issues related to allegations of leaking of product from Tank 16 andcircumstances surrounding an uplift event to the Tank 16 sump.

SITE VISIT

This writer visited the VMT terminal facility during the week of June 26, 2007. During the visitthe following activities were conducted:

(1) Observed the exterior of Tanks 16 and 55.

(2) Interviewed APSC personnel regarding the tank issues.

(3) Participated in hardness testing of the Tank 55 doorsheet welds.

(4) Reviewed radiographic film taken of the doorsheet welds.

The visit resulted in a request for production of additional documents pertinent to theinvestigation and those documents are detailed in Appendix A under APSC produceddocuments.

Hardness testing

At directed and witnessed by this writer, hardness tests of the doorsheet welds, heat-affectedzones (HAZ) and adjacent base metal on a sampling basis was conducted by others to addressthe expressed concerns with fitness-for-service of the Tank 55 doorsheet, i.e., unknown weldquality and brittle fracture. Based on the document review, others also believed that hardness


1Valdez Marine Terminal, Tank 55, Alleged Integrity Concerns Preliminary Investigation,March 15, 2007, Rev. 2, pg. 4 of 44.



Page 6 of 22

testing of the doorsheet welds would provide useful information; however, for unknown reasonsthe tests were never done.1

Hardness tests can be accurately correlated with the tensile strength of a ferrous metal/alloyand allow a qualitative estimate of the toughness of a material, i.e., as hardness increases, thetoughness of a ferrous metal/alloy decreases (although not linearly). As the HAZ can be thehardest and least tough location in a welded component, the hardnesses were conducted usinga Krautkramer MIC-10 microhardness test instrument, to permit measurements in this thin layer. Appendix B shows photographs of the tank, doorsheet, hardness test locations and testequipment. Hardness results for the Tank 55 doorsheet are detailed in Appendix C, Table C1.

The results showed that base metal, HAZ and weld metal hardness measurements were all low,indicating good ductility and toughness. The hardness measurements were slightly lowcompared with the minimum specified tensile strength for the base and weld metal. However,as they were compared after converting to their equivalent tensile strength (using acceptedASTM conversion factors) and the conversion doesn’t represent 100% equivalence, the slightlylow measurements are not considered an issue of importance.

Doorsheet Radiographic Film Review

After completion of welding on the Tank 55 doorsheet, the doorsheet welds were 100%radiographically inspected. This writer reviewed all film, including reshoots taken after minorweld repairs to eliminate porosity. The film reviewed showed that the doorsheet welds were ofhigh quality and met and exceeded ASME Section VIII, Paragraph UW-51(b) requirements, theapplicable code for the doorsheet welds.

TANK 55Tank Design and Contents

Summary design information associated with Tank 55 is detailed in table 1 below.




Page 7 of 22

table 1 - Tank 55 Design InformationSize 100' O.D. x 31'-8" ht.Capacity 40,000 Bbls.Service Diesel FuelSpecific Gravity 0.9Build Date 1976Specification API 650, App. GShell Material JIS SM41C.A. NoneMax. Product Ht. 28'-9"Weld Metal E7018-1

Base and weld metal properties

Drawing No. 73-7235-1A Rev. 7 for Tank 58-TK-55 indicates a notation offor plate material. This notation is found in document “CBI Material Identification forAlyeska Storage Tank,” REF-00122. It refers to Specification JIS G 3106 SM 41 C Mod. ClassA and states the following material properties in table 2.

table 2 - JIS SM41 Material Specification

Chemistry (wt.%)Carbon 0.18 max.Silicon 0.15-0.40Manganese 1.40 max.Phosphorus 0.035 max.Sulfur 0.30 max.

Processing Normalized and fine-grained. Mechanical Properties

Tensile Strength 58,000-73,000 psi.Yield Strength 34,000-58,000 psi

Notch Toughness 35 ft.-l:bs. Min. Average

Hardness 280 HV30 Max.




Page 8 of 22

Tank 55 Unsats

Background

PWSRCAC received a series of letters during May 2006 alleging faulty welds, incorrect use ofwelding procedures, and regulatory indifference pertaining to unspecified welding situations atthe Valdez Marine Terminal. The inquiries included welds that were performed on Tank 55(diesel fuel) in 2002. Ten welds were needed to reinstall the door sheet in Tank 55. Weldingwas reportedly started with an incorrect procedure, i.e., one for a crude oil tank having a shellthickness of 0.625 inches of steel. The shell thickness of Tank 55 is nominally 0.312 inches andthe range of thicknesses permitted by the procedure for welding crude oil storage tanks did notinclude the thickness for Tank 55. An inspector refused to accept the welds due to the incorrectprocedures. The remaining welds were completed with a valid (and accepted) set of weldingprocedures. The two sets of procedures specified the same welding variables except that thespecified heat input for the thinner shell was less than that permitted for the thicker shell. Thewelds remaining were then completed using the appropriate set procedures and appear to havebeen inspected and accepted without controversy. The non-conforming welds were not re-welded, as the in-process parameters would have allegedly been the same. Several welds inquestion were approved by a welding supervisor. Tank 55 was filled with diesel fuel shortlythereafter. It was reported that the non conforming Tank 55 welding procedures did notconsider the differences in maximum heat input of the questionable welds. The tank has beenthrough multiple fill and use cycles and three annual temperature cycles without leaking orshowing any other signs of failure.

Issues

The document review revealed several allegations associated with Tank 55 that are included inthe scope of this investigation (see reference in footnote 1).

1. Door sheet welded with wrong welding procedure

APSC, the agencies, and this report substantiated that the wrong welding procedure wasused while welding a portion of the Tank 55 door sheet; however, there was noconsensus on whether or not the use of the wrong welding procedure posed amechanical integrity issue that warranted additional testing or tank repair. APSCmaintained that the procedures were substantially similar, while the inspectors and Mr.Harrison with the JPO disagreed. (The JPO is a joint federal and state office with theFederal Bureau of Land Management and the Alaska Department of Natural Resourcesas the lead agencies).


2 Harvey Consulting LLC, Summary of Findings of Facts, Allegations andRecommendations, April 5, 2007.



Page 9 of 22

2. No heat input monitoring for door sheet weld

APSC’s quality assurance documentation substantiated that no heat input monitoringoccurred on about 60% of the Tank 55 door sheet welds. No quality control data wascollected on any of the exterior welds. No quality control data were collected on the twomain interior vertical welds. APSC maintained that monitoring data for 40% of the weldswas sufficient to ensure welding quality control. The inspectors maintained that APSC’sown welding procedures and quality control program required data to be collected on100% of the welds. Mr. Harrison (BLM) recommended a simple, cost-effective hardnesstest be completed on the heat-affected zone prior to returning the tank to service. Thistest was not done.

While APSC maintained that monitoring data for 40% of the welds was sufficient toensure welding quality control, the data appeared suspect to some. Even APSCengineers evaluating the data in 2006 agreed the data was abnormally consistent, anddid not reflect the variability one would expect for this type of welding process.Allegations of record tampering existed.

3. Welding out of sequence on door sheet

A record review confirmed the door sheet welding was completed out of sequence.

Known Facts and Assumptions

Based on information contained in a report by Harvey Consulting 2, the following is a summaryof substantiated and unsubstantiated facts and allegations associated with Tank 55 that arewithin the scope of this report:

Facts

1. A new tank floor was installed during the summer of 2002.

2. Tank 55 was returned to service on November 26, 2002.

3. ADEC and JPO did not provide a written record approving this tank to be returned toservice and there is no written agency finding on or before the tank was returned toservice resolving the allegations.




Page 10 of 22

Substantiated allegations (by Harvey Consulting)

1. The door sheet was welded with the wrong procedure and APSC did not demonstrateprocedural equivalence.

2. The door sheet monitoring data was suspect, and may have not been sufficient toensure quality control.

3. Alyeska’s quality documentation was not sufficient to demonstrate adherence to its owninspection, repair and quality control procedures regarding welding the door sheet out ofsequence.

Tank 55 UNSATS

WPS/PQR’s (UNSAT #19). There were three main allegations directed at Tank 55: (1)welding using an incorrect (and non-approved) WPS/PQR, (2) the doorsheet data monitoringdata was suspect, particularly related to weld heat input and, (3) out-of-sequence welding. However, based on a review of documents produced by PWSRCAC, it would appear thatAllegation No. 1 was considered the most egregious non-conforming item.

Below is an abbreviated chronology of the Tank 55 door sheet WPS/PQR allegation, asobtained from a Harvey Consulting, LLC report, Attachment B, March 15, 2007:

1998. TANCO and APSC developed two welding procedures6 for diesel storage tanks of0.312” thickness, and two welding procedures for crude oil storage tanks of a 0.625” shellthickness.

March 27, 2002. Mr. Stevens (APSC employee at this time) prepared a Tank 55 door sheetremoval, replacement, and welding/NDE/pressure testing procedure. This procedure wasapproved by Tony Balowski (APSC welding engineer). In error, the procedure called for use of acrude oil tank welding procedure rather than a diesel tank welding procedure.

September 14, 2002. The Tank 55 doorsheet was welded using welding procedures for a0.625”thick crude oil tank shell, instead of a thinner diesel tank shell. Welding continued throughSeptember 16, 2002 until the error was found. This error was documented in the Tank 55records as Unsatisfactory Finding No. 19 (UNSAT #19).

September 20, 2002. Inspector documents an Unsatisfactory Finding No. 20 (UNSAT #20),because an approved welding procedure was not in place for the Tank 55 when the welding ofthe door sheet occurred. APSC Welding Engineer amends the crude oil tank welding procedurespecifications to include diesel tanks such as Tank 55 four days after the welding procedureerror was found by Inspector Kale. KAM Inspector changes the Unsatisfactory InspectionFinding No.20 (UNSAT#20) to a satisfactory inspection finding.




Page 11 of 22

Appendix D, Table D1 contains a matrix of the original and revised ad resubmitted WeldingProcedure Specifications (WPS) and Procedure Qualification Records (PQR) used for the Tank55 doorsheet welding.

To distill the WPS/PQR issue:

A. Two welding procedures were originally issued for the doorsheet welding: (1) T-400-2G-1and T-400-3G-1. The 2G procedure was for the horizontal doorsheet welds and the 3G wasfor the vertical doorsheet welds. Both procedures were impact tested procedures with heatinput restrictions using E7018-1 SMAW welding electrodes. The original procedures T-400-3G-1Q PQR which was welding using a 0.625" coupon, which qualified the associatedwelding procedures for a 2T thickness range of 0.625"-1.25". Obvious, this thickness rangeexceeded the doorsheet thickness of 0.312"; therefore, did not meet API 653, API 650 orASME Section IX requirements.

B. To correct this non-conformance, APSC produced a PQR that had been pre qualified with awelded coupon thickness (0.280") appropriate for the doorsheet thickness (T-400-3G-6Q),associated it with the two original WPS’s, added the new thickness range to the WPS’s andreissued them with a new revision number. Other than the new PQR coupon thickness, allessential and supplemental essential variables were identical between the old and the newWPS and PQR, except that the maximum actual heat input for the revised PRQ (6Q) was40,500 J/in. instead of 46,800 J/in. in the original PQR (1Q).

Professional Opinion

It is the opinion of this writer, while the revised WPS/PQR was produced after the fact, it did notviolate any code requirements and did not invalidate the actual doorsheet welds as the essentialvariables associated with the revised PQR were identical to those of the initial PQR and met therequirements of the WPS. It is technically correct to reissue a WPS with a new revision no.when associated with a new PQR that does not violate ASME Section IX requirements.Therefore, this writer, if he had been in a responsible position regarding reviewing the doorsheetWPS/PQR non conformance, would have accepted the revisions and changed UNSAT #19 to asatisfactory inspection finding.

No heat input monitoring for doorsheet weld (UNSAT #20)

Background. On 9/20/02 a KAM Inspector listed an Unsatisfactory Condition No. 20 (UNSAT#20), based on requirement #6 in Work Order Package 32001526, which stated that theinspector must "verify that the preheat requirements of the WPS are followed and closelymonitored per the SRP TM 1 Welding Engineering Review and T-411." According the WPSTG-400-3G-1 (Rev. 5) and its associated Procedure Qualification Report (PQR) T-400-6G-1Q,heat input is a supplemental essential variable. However, the WTR does not have any records


3 WeldReport1.doc



Page 12 of 22

of the parameters for two vertical welds: FW-4 and FW-8. It is not indicated how the heat inputwas monitored for the vertical weld for TK 55.

In table 3 below is a table summarizing data from a Tanco Weld Tracking Report with ouraddition of the heat input in kJ/in. in the rightmost column. It shows that no data was collectedfor doorsheet welds FW-4 and FW-8. It also shows that the welding variables were, to quotefrom the reference in footnote 1, “abnormally consistent.”

Additional information 3 regarding the doorsheet heat input issue stated that:

The maximum heat input should have been 40,800 J/in. instead of 46,800 J/in. in the revisedWPS T-400-2G-1 (Rev. 5) and T-400-G-1 (Rev. 4) per ASME Section IX, QW-409.1 for the shellthickness range of 0.280-0.560 inches. The heat input for welds FW-1, FW-2, FW-3, FW-5,

table 3 - 58-TK-55 Diesel Storage TankWeld Tracking Report (WTR) (1)

Inspector James Mark HodgesNDE Type Radiographic Test

InsideWeldNo.

PreheatTemp (F)

WeldingSpeed

(in./min.)

Amps Volts A= AcceptR= Reject kJ/in

FW-1 165 6 120 23 A 27.6

FW-2 155 6 135 23 A 31

FW-3 165 6 125 23 A 28.7

FW-4 N/A N/A N/A N/A R -

FW-5 170 6 125 23 A 28.7

FW-6 165 6 135 23 R 31

FW-7 170 6 120 23 A 27.6

FW-8 N/A N/A N/A N/A A -

FW-9 155 6 135 23 A 31

FW-10 170 6 140 23 A 32.2

Note: No data for the corresponding outside welds.(1) WeldReport1.doc




Page 13 of 22

FW-6, FW-7, FW-9 and FW-10, as calculated from the in-process information documented inthe “abnormally consistent” Weld Tracking Report (WTR), ranged from a low of 27,600 J/in to ahigh of 32,200 J/in. No in-process welding variables (preheat, amperage, voltage, travel speed)were documented for the welding of the door sheet welds: FW-1A thru FW-10A (outside welds)and FW-4 and FW-8 (inside welds). Based on a review of the ASPC Surveillance/RepairProcedure TM-1, Rev. 1 (8/7/02) the extent of examination was to: “perform welding and visualin-process examination on a random basis.” Additionally, there is no requirement in API 650 orAPI 653 for in-process inspection of welding.


The heat input into a weld can be important or not, depending on many variables, including thehardenability of the alloy being welding, its thickness, it’s service requirements and distortioncontrol. The doorsheet welding procedures included maximum heat input as a supplementalvariable as the procedures were required to be impact tested. Too low heat input and too highheat input can detrimentally influence as-welded toughness properties; however, it is commonto specify a maximum heat input, as too high a heat input can degrade the HAZ of the weld,reducing its toughness. Everything else being equal, if the actual structure weld does notexceed the heat input of a Charpy tested welding procedure, then the weld should be expectedto exhibit similar toughness values.

The available project Weld Tracking Sheets (WTS) show that the doorsheet heat input, asmeasured by the welding voltage, current and travel speed were not monitored for severalinside welds and none of the outside welds. Also, the WTS data that is available is apparentlysuspect, based on its unusual consistency. If one chooses not to believe the data that isavailable, an argument could be made that it is not known whether any of the doorsheet weldsconformed to the maximum 40,800 J/in. or not. However, if one were to choose to believe theavailable data, then the heat input of the monitored inside welds did not exceed ~32,000 J/in.,well below 40,800 J/in. Based on the requirements of the APSC TM-1 Repair Procedure, repairexaminations were only required on an unspecified “random” basis, which the degree ofmonitoring, as represented by the available WTS met. Also, as correctly pointed out, there isno specific requirement in API 650 or API 653 for in-process inspection of welding addressingthis issue. Therefore, rather than approach the issue based on whether project requirementswere met or not, a more constructive approach would be to discuss: (a) the likelihood ofexceeding the maximum of 40,800 J/in. and, (b) if other, indirect evidence would suggestwhether the doorsheet welds are susceptible to brittle fracture or not.

This writer, for several reasons, believes that it is likely that the doorsheet welds did not exceed40,800 J/in. If one considers the three variables controlling heat input of a weld, i.e., voltage,current and travel speed, it is probably going to be the travel speed that is the most subject tovariation. The voltage and current are constrained by the diameter and weld metal transfercharacteristics of the electrode. In the monitored WTS inside weld data the voltage is aconstant 23 volts and the current ranges from 120 amps to 140 amps. These parametersclosely resemble the voltage and current used in the Lincoln Easy Arc 528 MR E7018-1Procedure Development Data Record from the Testing Institute of Alaska (11/28/94) that




Page 14 of 22

qualified the Lincoln 7018-1 electrode’ notch toughness data. In that data record, for anelectrode diameter of 1/8", the recorded volts for all passes ranged from 20.6- 22.8 volts and thecurrent from 118-122 amps. The travel speed ranged from 3.7-6.7 in. with all but two passesbeing greater than 4.8 in./min. Thus, one could conclude that the voltage and amperagerecorded on the WTS for the inside welds were probably similar. If so, then the most influentialvariable on heat input would be the travel speed. Based on a worse case combination of 140amps and 23 volts (table 3, FW-10) it would require that the travel speed be ~4.8 in./min. or lessto exceed 40,800 J./in. A normal weld speed for the doorsheet conditions would be more like8-9 in. min. and it’s not in a welder’s nature to maximize the time spent welding. Figures 6, 7and 8 in Appendix B show lengths of the doorsheet exterior weld cap pass. To this writer, theweld appearance does not suggest an excessively slow travel speed.

The above addresses the probability that the doorsheet welds conformed to the 40,800 J/in.requirement. Approached from another perspective, what is the probability that the doorsheetwelds are prone to brittle fracture? This argument is partly addressed by the doorsheet weldhardness measurements. The measurements are all very low, suggesting a ductile, tough weldand HAZ. Also, the radiograph review showed that all welds were free from injurious defectsthat could affect toughness. Therefore, it is this writer’s opinion that the doorsheet weldsprobably met the maximum 40,800 J/in. requirement, and that the weld quality is excellent andthat the material appears to be ductile and tough, based on the hardness data. So, while theheat input monitoring might be procedurally an issue, technically it is not.

Welding out of sequence on door sheet

Background. An unsat was written to because the door sheet welding was completed on theoutside of the tank (including corners and 12” cutbacks) before the inside horizontal corner and12” cutback welds were complete. After the outside of the doorsheet was completely welded,TANCO welded the inside horizontal joints and then the corners and cutbacks.

The sequence of welding that took place is as noted below (footnote 1):

1) outside vertical joints

2) inside vertical joints

3) outside horizontal joints (along with outside corners and cutbacks)

4) inside horizontal joints (along with inside corners and cutbacks)




Page 15 of 22

References

This writer found two APSC project document references that addressed the issue of doorsheetwelding sequence:

(1) Drawing D-54-TM1-C01, Tank Maintenance Typical Crude Tank Doorsheet Details, NoteNo. 6, “Prior to welding vertical cuts, cut at existing horizontal weld for a minimum of 12"beyond new vertical joints. Weld horizontal cuts last.”

(2) Project Technical Specification X052-T-411, Aboveground Storage Tank Repair andAlteration, paragraph. 3.4.6, Fit-up, Alignment, and Weld Sequence - “Use fit-up methodsand welding sequences for tank welding, welding bottom, shell, and roof plates thatminimize distortion due to weld shrinkage ....“Control weld distortion of large replacementplates by strictly adhering to fit-up procedures and weld placement sequences.

API 653 (December 2001) addresses sequence welding in paragraph 9.2.2.2, “Prior to weldingthe new vertical joints, the existing horizontal welds shall be cut for a minimum distance of 12 in.beyond the new vertical joints. The vertical joints shall be welded prior to welding the horizontaljoints.”


This writer believes that APSC technically met the project specification requirements and thoseof API 653, related to welding sequence. As suggested in APSC Project TechnicalSpecification X052-T-411, welding sequence is important for several reasons: (1) to controldistortion and, (2) to control excessive weld stresses that could lead to cracking. Based onobservation of the finished doorsheet during the site visit, there was no observable tank shelldistortion associated with the doorsheet replacement and the 100% magnetic particle andradiographic inspection verifies that the doorsheet welds are of good quality with no injuriousdefects.

Fitness-for-Service Discussion

Introduction to brittle fracture

Brittle fracture is fracture that involves little or no plastic deformation. It is usually associatedwith flaws or defects in the material where bulk stresses concentrate. A stress intensity isassociated with flaws or geometric notches and a stress concentration factor can be assigned tothe flaw or notch based on its geometry, location and orientation. The more acute the flaw ornotch, the greater the stress intensity.




Page 16 of 22

For a brittle fracture to occur in a normally ductile material, the following factors must be presentsimultaneously:

1. A stress concentrator must be present. This can be a weld defect, a fatigue crack, or ageometric notch such as a sharp corner, thread, hole, etc. The stress concentrator must belarge enough and sharp enough to be a "critical flaw" in terms of fracture mechanics.

2. A tensile stress must be present. The tensile stress must be of a magnitude high enough toprovide microscopic plastic deformation at the tip of the stress concentration. The tensilestress need not be an applied stress on the structure, but may be a residual stress insidethe structure, i.e., from welding or uneven cooling, etc.

3. The temperature must be relatively low for the steel concerned. The lower the temperaturefor a given steel, the greater the possibility that brittle fracture will occur. For some steelsthe ductile/brittle transition temperature may be above room temperature.

A fracture is "brittle" when it is associated with very little plastic deformation. Such fractures cantake place in otherwise ductile materials if they contain cracks. Brittle fractures have certaincharacteristics that permit them to be identified:

1. There is no gross permanent or plastic deformation of the metal in the region of brittlefracture.

2. The surface of a brittle fracture is perpendicular to the principle tensile stress.

3. Characteristic markings on the fracture surface frequently point back to the location fromwhich the fracture originated. These markings are sometimes referred to as "chevron" or"herringbone" marks.

Brittle fractures are dangerous because they occur suddenly, without warning (plasticdeformation), and the cracks travel rapidly. Equipment/components that are ductile can deformin the presence of a tensile stress and a flaw (crack) without fracturing. The vessel deforms(yields) instead of sustaining a rapid fracture. Brittle equipment/components can fracture atstresses below the yield strength of the material in the presence of a flaw and the cracks travelrapidly and without prior warning. This is what makes brittle fracture dangerous. In summary,brittle fracture can occur in a susceptible material (low toughness) under a tensile stress atwhich a “critical” size flaw is present. Implicit in this definition is that if a material containing aspecific size flaw has survived worse case conditions of the highest encountered stress andlowest toughness conditions then it should survive all conditions as severe and less severe inthe future. In the case of Tank 55 the worse case conditions would coincide with the highestproduct level (stress) at the lowest ambient temperature (toughness).

The most commonly used measure of a material’s toughness (resistance to brittle fracture) isthe Charpy impact test, typically reported in foot-pounds (ft.-lbs.). It is widely used as a


4 CBI material Identification for Alyeska Storage Tanks, REF-00122, Rev. 4/16/74



Page 17 of 22

measure of toughness because it is a relatively easy test to conduct. The ASME boiler andpressure vessel code (and API 650) recognized the importance of brittle fracture in theirspecifying a minimum level of toughness, historically 15 ft.-lbs. Today’s ASME and API codesaddress brittle fracture with “charpy-impact exemption curves,” which state the lowesttemperature a specific material can be used at a specified thickness.

Based on the low ambient temperatures that Tank 55 is exposed to in the winter season, thetank base metal and welds included required Charpy impact testing, in this case, 30 ft-lbs. at0oF. This toughness requirement was intended to insure that the tank base metal and weldscontained sufficient charpy-impact toughness to resist defects at the lowest expected ambienttemperature, i.e., 0F. Implicit in the code toughness philosophy is that the code mandated nondestructive inspection requirements will detect any defects that would be injurious to the vesselat the specified minimum charpy-impact toughness value.

Brittle Fracture Assessment to API 579, Section 3

The current “best practice” for conducting fitness-for-service assessments is API 579. API 579has procedures for assessing the potential for a brittle fracture, based on three, increasingcomplexity levels, i.e. Level 1, Level 2 and Level 3. Usually, a Level 3 brittle fractureassessment involves a fracture mechanics analysis. The Level 1 assessment is based oncomparing a Critical Exposure Temperature (CET) with a minimum allowable temperature(MAT). A Level 1 assessment passes if CET > MAT. A Level 2 assessment follows the right-hand path in the Appendix E figure.

Appendix E contains a logic diagram and supporting material extracted from API 579 used toassessment the potential for brittle fracture of an API 650 storage tank. In the Governing PlateThickness graph this writer believes that SM41 tank shell plate material belongs to Curve B (allnon-cast materials in Curve A if produced to fine-grained practice and normalized and not inCurves C and D). Based on the governing thickness for Tank 55 (0.312"), the charpy impactexemption temperature for the Tank 55 shell plate material is -20F. As the minimum designmetal temperature (MDMT) for Tank 55 is 0F, the original tank base metal construction passesa Level I brittle fracture assessment, i.e., (CET = 0oF > MAT = -20oF)

Based on fitness-for-service considerations, the question then becomes does the doorsheetreplacement fabrication pass a Level 1 assessment criteria. This writer believes that it does,based on the charpy-impacted T-400-6G-1Q PQR used to qualify doorsheet welding proceduresT-400-3G-1, Rev. 5 and T-400-2G-1, Rev. 4.

Tank 55 was constructed with SM41 carbon steel produced to a normalized and fine-grainedpractice4. The Alyeska Aboveground Storage Tank Repair and Alteration specification used forthe project call for a design toughness of 30 ft.-Lbs. at 0oF for weld metal of plate thickness 1/4"


5 Alyeska Pipeline Specification No. X052-T-411, Rev. 0.



Page 18 of 22

to 3/4" 5. The PQR was qualified using 10mm x 5mm charpy-impact test specimens (at -20oF). Based on the miminum 30 Ft.-Lbs. charpy-impact requirement, Table 2 (footnote 5) allows a0.67 reduction factor for 10mm x 5mm subsize charpy-impact test specimens for plate material1/4" to 3/8". This reduction factor corresponds to a required 20 Ft.-Lbs. minimum charpy-impactvalue for the PQR (30 x 0.67). The average impact value for the T-400-6G-1Q PQR weldmaterial was 25 Ft.-Lbs. and 39 Ft.-Lbs for the heat-affected zone. Also, no individual charpyvalue fell below 80% of the minimum average value requirement.

Based on the above logic, the original Tank 55 construction and the doorsheet replacementpass a Level 1 brittle fracture assessment, based on API 579. It is also worth noting that thetank also is considered safe from brittle fracture, based on two additional Level 2 pass/failcriteria, (see the right-hand path in the Appendix E figure). The tank shell course thickness isless than the 0.5" inch where API 579 considers brittle fracture to be an operative damagemechanism. Also the tank also normally operates below the 8 ksi tensile strength cut off where,at stresses below 8 ksi, experience has shown that there is insufficient applied stress to run abrittle crack (based on API 650 formulas, the Tank 55 shell at the bottom of the doorsheet doesnot exceed 8-ksi until the tank if filled to greater than 11 feet of diesel product). Liquid fill heightvs. ambient temperature data submitted by APSC for Tank 55 for the years 2004, 2006 and2007 show that the tank product height rarely exceeded 11 feet and the maximum fill heightreported was 14 feet.

Tank 55 Fitness-for-Service Opinion

Based on the above tank data and design information, the doorsheet replacement fabricationpasses a Level I brittle fracture assessment per API 579. The brittle fracture assessmentimplicitly assumes that the plate and weld material are free from injurious defects (greater thancode allowable) and that is the case here.

TANK 16

Tank Design and Contents

Summary design information associated with Tank 16 is detailed in table 4 below.




Page 19 of 22

table 4 - Tank 16 Design InformationSize 250' O.D. x 62'-3" ht.Capacity 510,000 Bbls.Service Crude OilSpecific Gravity 0.9Build Date 1976Specification API 650, App. GShell Material JIS SM50C.A. 1/8"Max. Product Ht. 58'-6"Weld Metal unknown

Base and weld metal properties

Drawing No. 1-2 Rev. 0 for Tank 58-TK-16 indicates notations of “1" and “2" for plate material.This notation is found in document “CBI Material Identification for Alyeska Storage Tank”, REF-00122. Material spec “1" is a specification for JIS G 3106 SM 50 C Mod. Class A and states thefollowing material properties in table 5.

table 5 - JIS SM50 Material SpecificationChemistry (wt.%)

Carbon 0.18 max.Silicon 0.15-0.40Manganese 1.50 max.Phosphorus 0.035 max.Sulfur 0.30 max.Niobium 0.05 max.Vanadium 0.08 max.

Processing fine-grained and normalized. Mechanical Properties

Tensile Strength 70,000-85,000 psi.Yield Strength 50,000-68,000 psi

Notch Toughness 35 ft.-l:bs. Min. AverageHardness 280 HV30 Max.


6 Valdez marine Terminal Tank 16, Alleged Integrity Concerns, Preliminary Investigation,Harvey Consulting, LLC, March 13, 2007.



Page 20 of 22

Background 6

Concerns originally arose during the 2002 Tank 16 inspection and repair work when employeesand inspectors noticed tank floor damage in the area of the sump that put the integrity of thetank floor in question. It was alleged that the floor buckling was so significant that the sump wasremoved. The floor was patched to cover the hole left after the sump was removed. Not onlywere there problems with the tank floor around the sump, but Tank 16’s floor had historicallybeen patched in a number of places, and required more patching in 2002, contributing to theintegrity concerns. Of particular concern was the potential for ground water and soilcontamination from the crude oil stored in Tank 16.

One submitted document, an e-mail from Bonnie Friedman to John Engles, dated “1011712002" states: “...a concerned employee came by my office to express concern about theissue of possible contamination in the West Tank Farm. This employee was doing tankwork in TK 16. Apparently water pressure under the tank was so high as to cause the sump tobe jacked up about 5 inches. The sump bottom was removed during the tank work and theemployee noted what he perceived as contaminated water under the sump area. He said thesump had been open for several days and the water was moving so he did not believe it wasfrom the inlet of the tank. Also, the tank bottom to 16 was found to be in good condition so hefelt it was unlikely that Tank 16 was the cause of the oil. The employee infers that there iscontaminated water in the tank farm within the gravel layer between the surface and the CBAliner.”

The below is an abbreviated chronology of events surrounding the Tank 16 floor issue, asabstracted from the reference in footnote 6:

• Summer and Fall 2002 - Tank 16 was cleaned, inspected, repaired (sump replaced due todeformation). It is not clear exactly when Tank 16 was returned to service, but it appearsthat it was sometime during the Fall of 2002.

• October 1, 2002 - Mr. Harrison (BLM) and Ms. Friedman (ADEC) met with a concernedemployee. The employee was working at the VMT and alleged that there might be a leak inTank 16’s floor. According to Mr. Harrison, Mr. Moore (APSC) obtained and testedsamples from the Tank 16 sump area, reporting no contamination. Mr. Harrison said he hadnot handled the sample or testing; this was all handled by ADEC and he never actually sawthe test results, but was told that the tests were negative for contamination by Ms.Friedman.




Page 21 of 22

• January 21, 2003 - An internal APSC memorandum from Ms. Lee (APSC) to Mr. Stokes(APSC) summarized the 2002 inspection findings for Tank 16. The report states that 80floor plate patches were installed on 49 of the original floor plates. There are 260 originaltank floor plates; 49 plate repairs out of 260 is approximately 19% of the total tank floor.The report confirms the entire sump was replaced due to deformation of some of the plates.

There was little documentation submitted regarding the above allegation and this apparentlyreflects the statement in the Harvey Tank 16 report, i.e., “ Very few sources of written recordswere available for this report.” No reviewed document verifies whether oil was actually leakingfrom under Tank 16 nor not.

Discussion

The Tank 16 sump documents suggests that the tank floor was repaired and “inspected”following the floor patch plates installation and sump removal. This writer could find nodocumentation that detailed exactly what inspection was conducted on the floor after therepairs. So a documentation review will not be sufficient to resolve the tank floor leaking issue.

To address the issue of Tank 16 leaking, this writer conducted a visual inspection of the exteriorof Tank 16 during the site visit. As part of the inspection, caps were removed from plastic pipeinspection ports placed below the tank floor around the circumference of the tank. When theinspection port caps were removed, water typically exited from the inspection port pipe. Closevisual observation and sniffing the water for hydrocarbons did not detect any tracehydrocarbons. In addition, various plants, grass and weeds, etc. around the periphery of thetank were healthy and green, also suggesting no exposure to hydrocarbons.


Based on the available information, it is not known whether Tank 16 ever leaked hydrocarbonsinto the ground; however, a visual inspection of the exterior of the tank showed no evidence ofexisting hydrocarbon leaks.

TANK 55 AND 16 CONCLUSIONS

• While certain irregularities occurred during installation of the doorsheet, those irregularitiesdid not influence the tank fitness for service, based on brittle fracture and weld integrityconcerns.

• While the original welding procedures approved for welding the Tank 55 doorsheet werenot in compliance with API Standard 650, Welded Steel Tanks for Oil Storage, and APSCproject specifications, revisions to the essential variables of the original procedures madeby APSC to correct the deficiencies (as documented by inspection UNSATS) were properand did not violate any code or project specification requirements.




Page 22 of 22

It is our policy to retain components and sample remnants for a minimum of 30 days from the report date, after which time they may be discarded. This report shall not be reproduced, except in full, without prior written permission from The Hendrix Group Inc.

SCOPE OF SERVICES: The agreement of The Hendrix Group Inc. to perform services extends only to those services provided for in writing. Underno circumstances shall such services extend beyond the performance of the requested services. IT is expressly understood that all descriptions,comments, and expressions of opinion reflect the opinions or observations of The Hendrix Group Inc. based on information and assumptions suppliedby the owner/operator and are not intended nor can they be construed as representations or warranties. The Hendrix Group Inc. is not assuming anyresponsibilities of the owner/operator and the owner/operator retains complete responsibility for the engineering, manufacture, repair and usedecisions asa result of the data or other information provided by The Hendrix Group Inc. In no event shall The Hendrix Group Inc.’s liability in respectof the services referred to herein exceed the amount paid for such services.

STANDARD OF CARE: In performing the services provided, The Hendrix Group Inc. uses the degree, care and skill ordinarily exercised under similarcircumstances by others performing such services in the same or similar locality. No other warranty, expressed or implied, is made or intended by TheHendrix Group Inc.

• Welding with initially incorrect welding procedures versus welding to the revisions made tocorrect those procedures would not have materially altered the actual welding process anddid not compromise the doorsheet welds.

• Issues regarding the doorsheet welding procedures and the actual welds made did notcompromise the original tank design.

• Review of the available documents did not permit an unqualified verification as to whetherproject specification and welding procedure requirements pertaining to heat inputrequirements were met; however, the discussion will show that, based on the availableinformation and the line of thought conducted, any deviations from the requirements did notcompromise the quality or fitness -for-service of the doorsheet welds.

• An API Recommended Practice 579, Fitness-for-Service, Level 1 and Level 2 assessmentregarding the potential for brittle fracture showed that, based on the available information,Tank 55 passed the Level 1 and 2 assessment and is considered fit-for-service, based onbrittle fracture concerns.

• It could not be verified whether Tank 16 ever leaked hydrocarbon during the 2002 timeperiod; however, an inspection of the exterior of the tank in June 2007 showed that it couldnot have been leaking significant quantities of product into the ground.

Appendix A

Indicates an electronic file folder.Page 1 of 5

REFERENCE MATERIAL

Provided by PWSRCAC

KakivikLitigation

kalevskakivikTSC4n502.pdfkalevskakivikTSC4n5.pdfvamvoras031204.pdfSarbaneOxley030304.pdfDOL030917.pdfkalevkakivik.pdf

May07BoardPack

ADECLetter030425.pdfTank93Report.pdfTank16Report.pdfTank5Report.pdfTank55Report.pdfFindingsSummary.pdfBriefingSheet.pdf

Misc

StaffFindingSummaryRev03a.docTank 55 - Weld Report1.docOutlineofVMTWeldReportTKEdits.docListQuestionsMikeStevensAlyeskaEdits060629.docListQuestionsTonyBalowskiAlyeskaEdits060629.doc6-26-02 Draft List of Questions for Tony Bilowski.doc6-26-02 Draft List of Questions for Mike Stevens.docWeldDocuments060620.docWeldMeetingsHandWrittenNotes0606.pdfWeldDocuments060614.docWeldNotes060601.PDF500.300.060510.HamelNSCtank.pdfChuckHammelGlenPlumlee.pdf

SlewisInquiry

WTRXO52Tank55DoorPg1.pdfWTRX052TankDOORPg2.pdfWTRX052ColumnPlatesPg1.pdfWTRTank55SUMPExt.pdf

of 5

SourceDocuments

DS012.PDF

ADEC Tank Docs

CD_1 Photo's 1 Photo's2 Photos 3 Tank 5 Tank 55

Photo's TK 55 TK 55 GUIDE CLIPS

Guide Clips p.1.docTK 55 SUMP WTR

TK. 55 NDE Sump attachments .xlsTank 55 SUMP Ext.docTK 55 WTR SUMP FLOOR PLATES.docTK 55 WTR SUMP piping.docTK. 55 NDE SUMP .xls

TK 55 WTR COLUMN PLATESWTR X052 Column Plates Pg. 1.doc

TK 55 WTR FLOORWTR X052 Tank 55 Floor page 1.docWTR X052 Tank 55 Floor page 2.docWTR X052 Tank 55 Floor page 3.doc

WTR ANNULAR RINGTK. 55 NDE ANULLAR RING .xlsWTR X052 Tank 55 Annular page 1.doc

WTR TK 55 DOORTK 55 LUG REPAIR

WTR XO52 LUG REMOVAL pg 2.docWTR XO52 LUG REMOVAL pg1.doc

NDE Request Tank 55 RT .xlsNDE Request Tank 55 weld repair.xlsTK. 55 NDE door sheet .xlsWTR XO52 Tank 55 Door Pg.1.docWTR XO52 Tank 55 Door Pg.2.docWTR XO52 Tank 55 Door Pg.3.docWTR XO52 Tank 55 Door Pg.4.doc

Tank 55 Arc Strike Repair Ext.doc

of 5

Tank 55 Ballard Ext.docTank 55 Shell nozzles (2).xlsTank 55.xnk

Tank 93photos 4README.TXTTank 16.xnkMisc E-mail.xnk

important docs

060510Hamel.PDF060516Hamel.PDF060518Hamel.PDF021011ADEC.PDF030404ADEC.PDF030425ADEC.PDF

JPO Info

503.300.031211.APSCbwtTanks.pdf500.300.030702.APSCcorMonit.pdf500.300.030702.APSCcpSystem.pdf500.300.030813.APSCbwtTanks.pdf500.300.031121.APSCtnkReqst.pdf500.300.060510.HAMELnsTank.pdf500.300.060512.APSCHamelCmt.pdf500.300.060516.HamelrespVMT.pdf500.300.060518.HAMELvmtTank.pdf500.400.060524.APSCweldInv.pdf503.300.031028.APSCreqInsp.pdf500.300.021220.APSCinspProg.pdf500.300.030129.APSC2002Acmp.pdf500.300.030131.APSCcrudeTnk.pdf500.300.030214.APSCtnkSched.pdf500.300.030306.ADECps12Tnks.pdf500.300.030318.APSCps12Tnk.pdf500.300.030410.APSCinspSchd.pdf500.300.030509.APSCweldActv.pdf030404ADEC.PDF030425ADEC.PDF500.107.020625.APSCweldBref.pdf500.300.020625.APSCweldProj.pdf500.300.020920.APSCtnk5Repr.pdf500.300.020925.APSCtnk55Inf.pdf500.300.021010.APSCcrudeIns.pdf500.300.021011.ADECtnkReprs.pdf500.300.021016.APSCtnkReprs.pdf

of 5

000128APSC.PDF000228APSC.PDF000308.APSC.PDF010319JPO.PDF010402APSC.PDF010418APSC.PDF020129APSC.PDF020620JPO.PDF020627ADEC.PDF020701APSC.PDF030324APSC.PDF

AlyeskaNewcomerCoverEmailForGJonesQandA.pdfRCAC QuestionsAnsweredByGregJones.docADEC Weld Report Cover Letter.pdfAPSC Controlled Report_2002 Valdez Tank Project Review final.pdf500.410.060703.APSCtnkRev02.pdf060512APSC.PDF

Reference Material Provided by Alyeska

Tank 55 documents requested and supplied by Alyeska

Column Guides.JPGDSC00034.JPGDSC00035.JPGDSC00038.JPGSKMBT_C35207073012050.pdfstillingwell1.JPGstillingwell2.JPGstillingwellroof2.JPGSump1.JPGTerminal Fuel Tanks Report (624) 2005.ZIP2004.XLS2006 (2).XLS2007.XLS

Hardcopy Documents

Tank 55DrawingsMaterial specifications of tank wallMTR of wall material - Unable to locateInspection reports that verify the tank doorsheet filler material - Unable to locateHydrotext exemption - Unable to locateFloor replacement drawingsTMPS of door sheetAPI 653 report

of 5

Welding sequence - Unable to locateProject report - X052-T411Inspection report - X052-T500IM manual/plan for ASTOriginal design calculations - Unable to locate

Appendix B


figure 1 - Tank 58-TK-55 at the APSC facility in Valdez, AK

figure 2 - 58-TK-55 tank nameplate. The contract number (73-7235) andbuild year (1976) are legible.


figure 3 - The taped areas represent where hardness measurements weretaken on the door sheet.

figure 4 - An example of a door sheet weld prepared for hardness testing.


figure 5 - Krautkramer Model KB-MIC 10 hardness tester used to take thehardness measurements (SN#4633).

figure 6 - Close-up of a doorsheet corner weld.


figure 7 - Close-up of a doorsheet horizontal weld.

figure 8 - Close-up of a doorsheet weld, possibly representing a repairweld.

Appendix C C

table C1 - Tank Doorsheet Hardness Measurements in 10Kg Vickers Units Thk. (In.)

LocationBM-

1HAZ-

1WM-CL

HAZ-2

BM-2

HAZ-LVT

HAZ-LVB

HAZ-RVT

HAZ-RVB

RWT

LVW 108 107 122 114 113 118 109 107 108 0.312.5

RVW 115 105 126 103 107 126 127 112 111 0.310

UHW 101 136 135 115 112 116 112 124 108 0.305

Notes:Hardnesses are an average of three readings or moreSM 41 T.S. = 58ksi-73ksi (122-156 Vickers)E7018-1 electrode = 70ksi min. (~150 Vickers)BM = base metalWM-CL = weld metal center lineHAZ = heat-affected zoneLVT = left vertical topLVB = left vertical bottom

Appendix D C

Table D1

Comparison of Welding Procedure Specification (WPS) Revisions in Accordance with ASME Section IX, QW-253

Applicable Essential TANCO WPS

or Supplementary T-400-2G-1 T-400-2G-1 T-400-3G-1 T-400-3G-1

Essential Variable Rev. 3 Rev. 4 Rev. 4 Rev. 5 Base Metal Group 2 1or2 2 1 or 2 Base Metal Thickness Range Impacts Minimum

.625" min. .625" min. Group 2 to Group 2 & .280" min. Group 1 or 2 to Group 1 or 2

.625" min. .625" min. Group 2 to Group 2 &. .280" min. Group 1 or 2 to Group 1 or 2

Base Metal Thickness Qualified

.625-1.25" Group 2 to Group 2

.625-1.25" Group 2 to Group 2 & .280-.560" Group 1 or 2 to Group 1 or 2

.625-1.25" Group 2 to Group 2

.625-1.25" Group 2 to Group 2 & .280-.560" Group 1 or 2 to Group 1 or 2

P-No. Qualified 1 1 1 1

F-Number 3 & 4 3 & 4 3 & 4 3 & 4

A-Number 1 1 1 1

Electrode Diam. > ¼ in. E6010: 1/8, 5/32 E7018-1:1/8 5/32, 3/16

E6010: 1/8, 5/32 E7018-l: 3/32, 1/8, 5/32, 3/16

E6010: 1/8, 5/32 E7018-l, 3/32, 1/8, 5/32

E6010: 1/8, 5/32 E7018-1: 3/32, 1/8, 5/32

AWS class. E6010/E7018-1 E6010/E7018-1 E6010/E7018-1 E6010/E7018-1 Deposit weld thickness 3/16" or less 3/16" or less 3/16" or less 3/16" or less

Position 2G 2G 3G 3G

Preheat 150°F for T<l"& 210°F for T<1”

150°F for T<l"&210°Ffor T<1”

150°F for T < l"&210°Ffor T<1”

150°F for T< l"&210°F for T<1”

PWHT N/A N/A N/A N/A

Increase in Heat Input 46,800 J/In. 46,800 J/In.* 46,800 J/In. 46,800 J/In.*

Current or polarity DC Reverse DC Reverse DC Reverse DC Reverse

• Should have been 40,800 J/In. for base metal thicknesses of 0.280-0.566-inches.

indicates change from previous WPS version

Appendix E C

Appendix F C

Listing of Acronyms

API - American Petroleum Institute

APSC - Alyeska Pipeline Service Company

ASME - American Society for Mechanical Engineers

ASTM - American Society for Testing Materials

BLM - Bureau of Land Management

FW - Field Weld

HAZ - Heat Affected Zone

JIS - Japanese Institute for Steel

JPO - Joint Pipeline Office

PQR - Procedure Qualification Record

PWSRCAC - Prince William Sound Regional Citizens Advisory Council

VMT - Valdez Marine Terminal

WPS - Welding Procedure Specification

WTR - Weld Tracking Report (same as WTS)

WTS - Weld Tracking Sheet

Prince William Sound Regional Citizens Advisory · PDF filePrince William Sound Regional...

Documents

Transcript of Prince William Sound Regional Citizens Advisory · PDF filePrince William Sound Regional...